The em dash became a popular signal for spotting AI-generated text. But as detection heuristics spread, models began adapting, exposing the fragility of stylometric fingerprints for synthetic content detection.
A deep dive into diffusion models—the core technology behind modern AI image and video generation—explaining how systems learn to transform random noise into coherent visuals through iterative denoising.
YouTube unveiled a sweeping suite of generative AI tools at its Made on YouTube event, from AI-generated thumbnails and edits to conversational creation assistants that automate much of the video production workflow.
A new research approach called Mean Velocity Matching reframes how diffusion models learn generative dynamics, promising more efficient sampling and higher-quality synthesis for the image and video generators driving modern synthetic media.
A new sim-to-real study finds that LLM synthetic personas don't reliably predict real audience responses to ad copy—and a simpler no-persona baseline actually beats persona-based simulation. A cautionary result for synthetic audience research.
Snorkel AI has tripled its valuation to $3.5B as enterprise demand for high-quality AI training data surges. The programmatic data-labeling firm is riding a wave of investment in the infrastructure that powers modern generative and multimodal models.
A UN scientific panel urges governments to adopt precautionary AI safeguards rather than wait for definitive proof of harm — a stance with major implications for deepfake and synthetic media regulation worldwide.
New research explores how large language models generate scenario-based fake news through discourse structures — and how detectors can be trained to catch it. A closer look at the arms race between synthetic text generation and authenticity.
AMD's move toward Taalas-style hardwired AI inference silicon challenges the GPU's dominance. Here's why custom chips baking models directly into silicon could reshape the economics of running generative AI — including video and voice models.
A new benchmark systematically tests whether large language models comply with China's AI-generated content regulations, revealing gaps in how models handle labeling, safety, and legal requirements for synthetic media.
New research reveals that text watermarking schemes designed to identify AI-generated content can inadvertently make language models more susceptible to adversarial prompt attacks, exposing a critical tension between authenticity tooling and model security.
Model collapse is the slow-motion degradation that happens when AI models train on AI-generated data. As synthetic media floods the web, future models risk losing diversity, accuracy, and quality in a compounding feedback loop.